INDEX

Explanations

information about uncertainty and confusion regarding identity or events

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Question words followed by additional context

np_acts-logits-general · gemini-2.0-flash

about who is acting

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nahilalakip

-0.66

 الحره

-0.63

 Италијани

-0.55

Datuak

-0.55

θηκαν

-0.53

addContainerGap

-0.52

 guste

-0.52

RenderAtEndOf

-0.51

ьаж

-0.50

fjspx

-0.49

POSITIVE LOGITS

who

2.28

Who

1.97

who

1.94

Who

1.94

 siapa

1.87

 hvem

1.80

 quién

1.74

 Siapa

1.62

 quien

1.53

誰が

1.52

Activations Density 0.356%

information about uncertainty and confusion regarding identity or events

Question words followed by additional context

about who is acting

No Comments

No Known Activations